layoutlm-sroie_synthetic

This model is a fine-tuned version of microsoft/layoutlm-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1204
• Ate: {'precision': 0.7636363636363637, 'recall': 0.8571428571428571, 'f1': 0.8076923076923076, 'number': 49}
• Ddress: {'precision': 0.8653846153846154, 'recall': 0.9, 'f1': 0.8823529411764707, 'number': 50}
• Ompany: {'precision': 0.9230769230769231, 'recall': 0.96, 'f1': 0.9411764705882353, 'number': 50}
• Otal: {'precision': 0.9215686274509803, 'recall': 0.94, 'f1': 0.9306930693069307, 'number': 50}
• Overall Precision: 0.8667
• Overall Recall: 0.9146
• Overall F1: 0.8900
• Overall Accuracy: 0.9778

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 3e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 4
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Ate	Ddress	Ompany	Otal	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.4783	1.0	43	0.1302	{'precision': 0.72, 'recall': 0.7346938775510204, 'f1': 0.7272727272727272, 'number': 49}	{'precision': 0.8113207547169812, 'recall': 0.86, 'f1': 0.8349514563106797, 'number': 50}	{'precision': 0.9423076923076923, 'recall': 0.98, 'f1': 0.9607843137254902, 'number': 50}	{'precision': 0.7142857142857143, 'recall': 0.8, 'f1': 0.7547169811320756, 'number': 50}	0.7962	0.8442	0.8195	0.9682
0.0285	2.0	86	0.1126	{'precision': 0.75, 'recall': 0.8571428571428571, 'f1': 0.7999999999999999, 'number': 49}	{'precision': 0.7962962962962963, 'recall': 0.86, 'f1': 0.826923076923077, 'number': 50}	{'precision': 0.9230769230769231, 'recall': 0.96, 'f1': 0.9411764705882353, 'number': 50}	{'precision': 0.9215686274509803, 'recall': 0.94, 'f1': 0.9306930693069307, 'number': 50}	0.8451	0.9045	0.8738	0.9764
0.0119	3.0	129	0.1177	{'precision': 0.7962962962962963, 'recall': 0.8775510204081632, 'f1': 0.8349514563106796, 'number': 49}	{'precision': 0.8846153846153846, 'recall': 0.92, 'f1': 0.9019607843137256, 'number': 50}	{'precision': 0.9423076923076923, 'recall': 0.98, 'f1': 0.9607843137254902, 'number': 50}	{'precision': 0.9215686274509803, 'recall': 0.94, 'f1': 0.9306930693069307, 'number': 50}	0.8852	0.9296	0.9069	0.9797
0.0078	4.0	172	0.1204	{'precision': 0.7636363636363637, 'recall': 0.8571428571428571, 'f1': 0.8076923076923076, 'number': 49}	{'precision': 0.8653846153846154, 'recall': 0.9, 'f1': 0.8823529411764707, 'number': 50}	{'precision': 0.9230769230769231, 'recall': 0.96, 'f1': 0.9411764705882353, 'number': 50}	{'precision': 0.9215686274509803, 'recall': 0.94, 'f1': 0.9306930693069307, 'number': 50}	0.8667	0.9146	0.8900	0.9778

Framework versions

• Transformers 4.50.0
• Pytorch 2.1.0+cu118
• Datasets 3.4.1
• Tokenizers 0.21.1